The AI talent war just escalated again

Let’s unpack what’s new…

OpenAI co-founder Andrej Karpathy has joined Anthropic’s pre-training team, putting one of the industry’s most influential AI researchers at the center of Claude’s future development. Karpathy will focus on using AI itself to accelerate model research, a strong signal that Anthropic sees AI-assisted R&D as a key advantage in the race against OpenAI and Google.

Karpathy is one of the few people with deep experience across both frontier model research and large-scale training systems, having helped shape early OpenAI research before later leading Tesla’s Autopilot and Full Self-Driving efforts. His return to frontier lab work also comes as Anthropic continues aggressively expanding its research and security teams.

The bigger shift is that frontier labs are increasingly competing on research velocity, not just compute. The next phase of the AI race may depend less on who buys the most GPUs and more on who can use AI to speed up discovery, experimentation, and model improvement itself.

Supported by Outskill

Have you been noticing how Claude is the most talked about tool on the internet right now? Claude has been launching new models & features like Claude co-work, Claude design, skills, connectors literally every week.

The World’s first Claude-a-thon, A 2 day deep dive into Claude, its user-cases and 10+ other AI tools happening this weekend from 10 am-7pm EST – and they have just 1000 free seats for a limited time.

In the workshop, you’ll do deep research on Claude. Build your own artifacts and dashboards. Create full presentations using Claude. Set up Claude Connectors like Indeed to automate your job search.

Register NOW! (free for the next 48 hrs)

Standard Chartered plans to cut nearly 8,000 roles as it pushes AI deeper into the business, with CEO Bill Winters explicitly saying the bank is replacing “lower-value human capital” with machines. The cuts will mainly hit back-office functions like HR, risk, and compliance as the bank tries to boost productivity and reduce operational costs.

What makes this notable isn’t just the layoffs, it’s how openly executives are framing them. For years, large companies positioned AI as a productivity assistant. Now major financial institutions are increasingly describing AI as a direct workforce replacement strategy tied to profitability and efficiency targets.

The bigger signal is that AI adoption inside enterprise is moving from experimentation into organisational redesign. Banks, consulting firms, and large corporates are no longer asking whether AI can reduce headcount. They’re starting to build long-term operating models around the assumption that it will.

Amazon’s Trainium chips are beginning to win over AI developers who previously defaulted to Nvidia, helped by improving software support and the ongoing shortage of Nvidia GPUs. Some companies are already shifting inference workloads to Trainium after seeing meaningful cost reductions compared to Nvidia’s H100 chips, while Amazon says its custom silicon business could eventually generate $50 billion annually.

The bigger signal is that the AI infrastructure market is slowly becoming less Nvidia-dependent. Nvidia still dominates frontier model training, but as AI workloads diversify, especially around inference and cost-sensitive deployments, developers are becoming more willing to adopt alternatives if the economics are compelling enough.

Looking to promote your company, product, service, or event to 100,000+ AI leaders?